JP5218544B2 - E-book trial reading data creation device and creation method - Google Patents

Description

  The present invention relates to an electronic book trial reading data creation apparatus and creation method, and more particularly to a technique for creating trial reading data for an electronic book provided using a network.

  In recent years, there has been a wide variety of terminal devices that can browse electronic books, and users can use various hardware devices such as personal computer, dedicated book readers, tablet terminal devices, and smartphones. You can browse books. In response to such enhancement of the hardware environment, sales of electronic books, which are software, are also increasing rapidly. In general, the price of e-books is set lower than the price of paper-based books, and because it has the advantage that it can be easily obtained using the Internet, it is expected that it will continue to grow in the future. .

  However, in the case of paper-based books, users can take the steps of actually picking up the books they are interested in at the bookstore, reading them, examining them, and purchasing them. In the case of an electronic book, it is difficult to provide a user with a free trial reading service in the same form. For example, if a service that allows users to freely read the contents of all pages of an electronic book via the Internet is provided, the user can browse all pages for free without any restrictions. The result will be the same as if the book was provided free of charge. However, there is a strong demand from users who purchase electronic books that they want to purchase after confirming their contents.

  Therefore, for example, Patent Document 1 below discloses a technique for encrypting a part of an electronic book and selling a browsing key for allowing browsing of the encrypted part for a fee. Patent Documents 2 and 3 disclose techniques that allow a user to read a trial of an electronic book after imposing restrictions on the pages and time that can be browsed. Furthermore, Patent Document 4 discloses a technique for preparing sample reading data for each electronic book in advance and providing the sample reading data to the user free of charge.

JP 2000-1113050 A JP 2003-122968 A JP 2003-150833 A JP 2003-296593 A

  As disclosed in some of the above-mentioned patent documents, if a part of the contents of an electronic book is extracted in advance as trial reading data, the trial reading data can be provided to the user free of charge via the Internet. A trial reading service similar to browsing at a bookstore can be provided for electronic books. In fact, a major e-book sales company has already provided such a sample reading service, and the user obtains the sample e-reading data of the desired e-book free of charge via the Internet and examines the contents. The electronic book can be purchased.

  However, the general method of creating trial reading data currently being performed is such as “extract the first few pages” or “extract the first few chapters” from the original electronic book data. Since a mechanical extraction method is employed, appropriate sample reading data is not necessarily created from the viewpoint of “inviting users to purchase”. For example, in the case of a story in which the story dynamically develops after the second half of the book, there is a high possibility that many users will find “interesting” in the sample reading data created by the conventional method. That is, the portion presented to the user as the trial reading data is not necessarily a portion that many users feel “interesting”, and therefore, the effect of causing the user who finished the trial reading to purchase the book is not sufficient.

  Of course, it would be ideal if the staff in charge of sales promotion carefully read the contents one by one and manually extract the parts that seem to be effective for inviting users to purchase, and create sample reading data. Trial reading data could be created. However, in the future, a large number of books are expected to be published as electronic books, and it is extremely inefficient to create sample reading data manually for each of these books.

  Therefore, an object of the present invention is to provide an electronic book sample reading data creation device and a creation method capable of efficiently creating sample reading data effective for inviting the user to purchase.

(1) According to a first aspect of the present invention, in an electronic book trial reading data creation device that creates trial reading data for an electronic book provided using a network,
A terminal device and a server device connected to the network are provided,
The terminal device
An electronic book data storage unit for storing electronic book data distributed from the server device;
A display that displays the contents of the e-book;
An instruction to receive an instruction input including an instruction to select a specific electronic book as a browsing target book or an instruction to cancel the selection and an instruction to specify a specific part of the selected electronic book as a browsing target part. An input section;
Based on the instruction input, the browsing control unit that reads data related to the browsing target part of the browsing target book from the electronic book data storage unit, and displays the contents of the electronic book on the display based on the read data;
A reading log collecting unit that collects a part or all of the history of instruction input as a reading log and transmits this to a server device via a network;
Have
Server device
A distribution target data storage unit that stores electronic book data to be distributed, and distributes it to the terminal device as necessary;
A reading log storage unit for storing a reading log transmitted from the reading log collection unit;
A reading log extraction unit that extracts a reading log related to a creation target book for which trial reading data is to be created from a reading log storage unit;
By analyzing the extracted reading log, the degree of interest indicating the degree of interest of the user for each part of the creation target book is obtained, and the degree of interest for creating the interest degree distribution indicating the distribution of the degree of interest in parts. A distribution creation unit;
A high interest region extraction unit that extracts a high interest region where the interest level is equal to or higher than a predetermined threshold based on the interest distribution;
Based on the electronic book data of the creation target book stored in the distribution target data storage unit, the trial reading that creates the trial reading data that includes the first half of the data but not the second half of the data of at least one high interest area. A data creation unit;
It is made to have.

(2) According to a second aspect of the present invention, in the electronic book trial reading data creation device according to the first aspect described above,
Provide multiple terminal devices,
The reading log storage unit stores reading logs of a plurality of users transmitted from each terminal device,
The reading log extraction unit extracts reading logs of a plurality of users related to the creation target book,
The interest level distribution creating unit creates the interest level distribution by analyzing reading logs of a plurality of users.

(3) According to a third aspect of the present invention, in the electronic book sample reading data creation device according to the first or second aspect described above,
The electronic book data is composed of a collection of page unit data,
The instruction input unit accepts an instruction input including an instruction to specify a browsing target part on a page,
The browsing control unit displays the contents of the specified page on the display,
The reading log collection unit collects the display history of the specified page as a reading log,
The interest level distribution creation unit creates the interest level distribution indicating the distribution of the interest level in units of pages by obtaining the interest level for each page,
The high interest area extraction unit extracts a high interest area as an area having a page as a minimum unit.

(4) According to a fourth aspect of the present invention, in the electronic book sample reading data creation device according to the first to third aspects described above,
The interest distribution creation unit recognizes the browsing target part at the time of deselecting the electronic book as the browsing suspended part of the electronic book, and for each part of the electronic book, shows the cumulative number of interruptions indicating the number of times it became the browsing suspended part. Thus, the cumulative number of interruptions is counted as a negative interest level to create an interest level distribution.

(5) According to a fifth aspect of the present invention, in the electronic book sample reading data creation device according to the fourth aspect described above,
The high interest area extraction unit does not extract the leading area of the electronic book as a high interest area.

(6) According to a sixth aspect of the present invention, in the electronic book sample reading data creation device according to the fourth or fifth aspect described above,
When the instruction input unit is given an instruction to close the electronic book data file, the instruction input unit recognizes that the instruction to cancel the selection of the electronic book is given,
The reading log collection unit collects the browsing target part at the time when the instruction is given as a reading log indicating the browsing interruption part for the electronic book.

(7) According to a seventh aspect of the present invention, in the electronic book sample reading data creation device according to the fourth to sixth aspects described above,
When the instruction input unit is instructed to put the “terminal device itself” or “electronic book browsing function of the terminal device” in a standby state or stopped state while browsing the electronic book, the selection of the electronic book is canceled. Recognizing that the instructions were given,
The reading log collection unit collects the browsing target part at the time when the instruction is given as a reading log indicating the browsing interruption part for the electronic book.

(8) According to an eighth aspect of the present invention, in the electronic book sample reading data creation device according to the fourth to seventh aspects described above,
The reading log collecting unit collects the reading time as the reading log together with the instruction to specify the browsing target part.
The interest distribution generation unit recognizes the display time from which the same browsing target part has been continuously displayed from the reading log, and if the display time exceeds a predetermined allowable range, the browsing target part is browsed. It is handled as an interrupted part.

(9) According to a ninth aspect of the present invention, in the electronic book sample reading data creation device according to the first to third aspects described above,
The interest level distribution creation unit creates the interest level distribution by counting the cumulative number of displays for each part of the creation target book as a positive interest level.

(10) According to a tenth aspect of the present invention, in the electronic book sample reading data creation device according to the first to third aspects described above,
The reading log collecting unit collects the reading time as the reading log together with the instruction to specify the browsing target part.
The interest level distribution creation unit recognizes the display time of each browsing target part from the reading log, and counts the integrated display time for each part of the created target book as a positive interest level to create an interest level distribution It is what I did.

(11) According to an eleventh aspect of the present invention, in the electronic book sample reading data creation device according to the tenth aspect described above,
The reading log storage unit stores the reading log separately for each user,
The interest distribution creation unit normalizes the display time for each part for each user, and then counts the integrated value of the normalized display time as a positive interest degree to create an interest distribution. It is.

(12) According to a twelfth aspect of the present invention, in the electronic book trial reading data creation device according to the tenth or eleventh aspect described above,
When the display time during which the same browsing target part is continuously displayed exceeds a predetermined allowable range, the interest level distribution creation unit does not include the display time in the integrated value.

(13) According to a thirteenth aspect of the present invention, in the electronic book sample reading data creation device according to the first to third aspects described above,
The instruction input unit has a function of accepting an instruction input for adding auxiliary information to a specific portion of the selected electronic book,
The browsing control unit stores auxiliary information, and has a function of adding and displaying auxiliary information when displaying a specific location on the display,
The reading log collection unit collects a reading log including an instruction to add auxiliary information, and transmits this to the server device via the network.
The interest level distribution creating unit counts the addition frequency of the auxiliary information for each part of the creation target book as a positive interest level to create the interest level distribution.

(14) According to a fourteenth aspect of the present invention, in the electronic book trial reading data creation device according to the thirteenth aspect described above,
As auxiliary information, various marks, underlines, or memos are added.

(15) According to a fifteenth aspect of the present invention, in the electronic book sample reading data creation device according to the first to fourteenth aspects described above,
The trial reading data creation unit sets a division position at an arbitrary position in one high interest level area, and uses data from the top of the high interest area to the division position as trial reading data.

(16) According to a sixteenth aspect of the present invention, in the electronic book sample reading data creation device according to the first to fourteenth aspects described above,
The trial reading data creation unit sets a division position at an arbitrary position in one high interest level region, and uses data up to a position retroactive by a predetermined amount from the division position as trial reading data.

(17) According to a seventeenth aspect of the present invention, in the electronic book sample reading data creation device according to the first to fourteenth aspects described above,
The trial reading data creation unit sets the division position at an arbitrary position in one high interest area, and the data from the division position to the position that goes back to the beginning of the paragraph, section, chapter, or the entire book including the division position. Test reading data is used.

(18) According to an eighteenth aspect of the present invention, in the electronic book sample reading data creation device according to the first to fourteenth aspects described above,
The trial reading data creation unit sets a division position at an arbitrary position in each high interest level area, divides each high interest level area into a first half part and a second half part with the set division position as a boundary, and distribution target data Data obtained by removing the latter half of each high interest area from the electronic book data of one book stored in the storage unit is used as the trial reading data.

(19) According to a nineteenth aspect of the present invention, in the electronic book trial reading data creation method for creating trial reading data for an electronic book provided using a network,
Based on the user's instruction input, the terminal device having a function of displaying the contents of the electronic book on the display and allowing the user to browse the contents collects the user's browsing history as a reading log, and this is collected via the network. A reading log sending stage to send,
A reading log reception stage in which a server device that creates trial reading data receives a reading log transmitted from a plurality of terminal devices via a network;
The server device analyzes the received reading log to obtain the interest level indicating the degree of interest of the user for each part of the specific electronic book, and creates the interest level distribution for the specific electronic book The creation stage,
A high interest level region extraction stage in which the server device sets a threshold value for the interest level of the interest level distribution and extracts a high interest level region in which the interest level is equal to or greater than the threshold value;
Based on the electronic book data of a specific electronic book, the server device performs the process of extracting the data of the first half part from the high interest area or the process of deleting the data of the second half part of the high interest area. The trial reading data creation stage to be created,
Is to do.

(20) In a twentieth aspect of the present invention, in the electronic book trial reading data creation method according to the nineteenth aspect described above,
By analyzing the reading log in the interest distribution creation stage, the user can recognize the browsing interruption part that is estimated to have interrupted the browsing, and for each part of the electronic book, the cumulative number of interruptions that became the browsing interruption part The interest degree distribution is created by counting the negative interest degree.

(21) According to a twenty-first aspect of the present invention, in the electronic book trial reading data creation method according to the nineteenth aspect described above,
By analyzing the reading log at the stage of creating the interest level distribution, the cumulative number of displays for each part, the accumulated value of the display time, or the frequency of auxiliary information added by the user is counted as a positive interest level, and the interest level distribution Is to create.

  According to the present invention, when a user browses an electronic book using a terminal device, the browsing history is collected as a reading log and transmitted to a server device via a network. In the server device, an interest degree distribution indicating the distribution of the interest degree of the user is created for each part (for example, individual page) of each book based on the reading logs of a large number of users, and the interest degree is a predetermined threshold. A region that is equal to or greater than the value is extracted as a high interest region. Then, trial reading data is created from the data in the first half of the high interest area.

  The high interest level region is a portion of the book that is estimated to have been felt “interesting” by many users. The sample reading data created according to the present invention is configured to include at least one high interest area including the data of the first half but not the data of the second half. Will cause an urge to read the continuation (second half). For this reason, it is possible to create trial reading data effective for inviting the user to purchase.

  Moreover, the sample reading data according to the present invention is created by automatic processing of the server device based on reading logs of a large number of users collected via the network. Therefore, no manual work is required, and even when a large number of books are distributed as electronic books on the Internet, it is possible to efficiently create test reading data efficiently. In other words, if a user who has already purchased an e-book or a user who has applied for a monitor is requested to cooperate, a reading log is automatically collected if these users normally perform an e-book browsing act, Since the trial reading data is created based on these reading logs, appropriate trial reading data can be automatically created by a very rational method.

  In general, reading is interrupted at several points along the way before reading a book. In general, the user tends to read the portion that is considered “interesting” at a stroke and interrupt reading at the portion where the “interesting” has dropped. Therefore, by analyzing the reading log and recognizing the place where reading was interrupted and giving a negative interest level to the interrupting place, the interest level of the place where many users are likely to interrupt reading. On the contrary, it is possible to extract a portion where many users are difficult to interrupt reading (that is, a portion where many users feel “interesting”) as a high interest level region.

  Also, the part that the user repeatedly read many times (the part with a high cumulative display count), the part that took time to read (the part with a long display time), the part with auxiliary information (the part with the underline added) If a positive interest level is given to the user, it is possible to extract a portion in which the user is interested as a high interest level region.

It is a block diagram which shows the basic composition of the trial reading data preparation apparatus of the electronic book which concerns on this invention. It is a figure which shows an example of the browsing state of the electronic book in the terminal device 100 shown in FIG. It is a figure which shows an example of the user's instruction | indication input which the instruction | indication input part 140 shown in FIG. 1 takes in. It is a figure which shows an example of the reading log which the reading log collection part 150 shown in FIG. 1 collects. It is a figure which shows the browsing interruption part which the interest degree distribution preparation part 240 shown in FIG. 1 recognizes based on the reading log shown in FIG. It is a figure which shows an example of the high interest level area | region extracted by the high interest level area | region extraction part 250 based on the graph of the interest degree distribution created by the interest degree distribution creation part 240 shown in FIG. It is a graph which shows an example of the process in which the interest level distribution creation part 240 shown in FIG. 1 normalizes display time for every user. It is a figure which shows another example of the high interest level area | region extracted by the graph of the interest degree distribution created by the interest degree distribution creation part 240 shown in FIG. It is a figure which shows some examples of the sample reading data preparation process by the sample reading data preparation part 260 shown in FIG. It is a flowchart which shows the basic procedure of the trial reading data preparation method of the electronic book which concerns on this invention.

  Hereinafter, the present invention will be described based on the illustrated embodiments.

<<< §1. Basic configuration of sample reading data creation device >>>
FIG. 1 is a block diagram showing a basic configuration of an electronic book trial reading data creation apparatus according to the present invention. This device is a device for creating sample reading data for an electronic book provided using a network, and is configured by a terminal device 100 and a server device 200 as shown in the figure. Here, both the terminal device 100 and the server device 200 are connected to a network 300 (in this example, the Internet) and can exchange information with each other.

  The terminal device 100 is a device used by a user for browsing an electronic book. As shown in the upper part of the figure, the terminal device 100 is an electronic book data storage unit 110, a browsing control unit 120, a display 130, an instruction input unit 140, and a reading log collection unit 150. It has.

  Here, the electronic book data storage unit 110 has a function of storing electronic book data distributed from the server device 200 via the network 300. Specifically, an electronic book includes a storage device such as a hard disk drive, an SSD (Solid State Drive), and a semiconductor memory, and an input / output device that performs processing of downloading electronic book data via the network 300 and storing the downloaded electronic book data in the storage device. The data storage unit 110 can be configured. The figure shows a state where electronic book data files F1, F2, and F3 are stored. Here, for convenience of explanation, electronic books whose contents are stored as files F1, F2, and F3 are also referred to as electronic books F1, F2, and F3 using the same reference numerals.

  The contents of the electronic book stored in the electronic book data storage unit 110 are read by the browsing control unit 120 and displayed on the display 130. Which part of which electronic book is to be displayed is determined based on the user's instruction input. Here, an electronic book selected as a browsing target by the user is referred to as a “browsing target book”, and a specific portion designated by the user in the “browsing target book” is referred to as a “browsing target part”.

  The instruction input unit 140 receives an instruction to select a specific electronic book as a browsing target book or an instruction to cancel the selection, and an instruction to specify a specific part of the selected electronic book as a browsing target part. It fulfills the function of accepting instruction input. Based on this instruction input, the browsing control unit 120 reads data related to the browsing target part of the browsing target book from the electronic book data storage unit 110 and displays the contents of the electronic book on the display 130 based on the read data. I do.

  On the other hand, the reading log collection unit 150 collects a part or all of the history of the instruction input captured by the instruction input unit 140 as a reading log, and performs processing of transmitting the history to the server device 200 via the network 300. Which part of the instruction input captured by the instruction input unit 140 is collected as the reading log may be appropriately determined in consideration of processing performed on the server device 200 side described later. More specifically, information necessary for processing of the interest level distribution creation unit 240 may be collected as a reading log.

  The terminal device 100 shown in FIG. 1 can be configured as a dedicated book reader, or can be configured by incorporating a dedicated application program in a personal computer, a tablet terminal device, a smartphone, or the like. Among the constituent elements of the terminal device 100, the electronic book data storage unit 110, the browsing control unit 120, the display 130, and the instruction input unit 140 are constituent elements included in a general terminal device having an electronic book browsing function. These components themselves are known ones. The terminal device 100 used in the present invention is characterized in that a reading log collection unit 150 is further added to these components. After all, the terminal device 100 used in the present invention is configured by incorporating a dedicated program having a processing function as the reading log collection unit 150 into a general terminal device having a conventional electronic book browsing function. Can do.

  Here, when the terminal device 100 is configured using a personal computer, a tablet terminal device, a smartphone, or the like, the downloaded electronic book data files F1, F2, and F3 are usually stored in a hard disk drive or an SSD. In this case, when the instruction input unit 140 receives a command input to open a specific file, the data of the specific file is expanded on a memory such as a RAM, and a part of the data is displayed on the display 130. I will take it. The electronic book data storage unit 110 shown in FIG. 1 indicates a data storage location including all of the hard disk drive, SSD, RAM memory, and the like in such a case.

  In FIG. 1, only one terminal device 100 is shown for convenience, but in practice, it is preferable to use a plurality of terminal devices 100 when implementing the present invention. As described above, the terminal device 100 used in the present invention has a processing function as the reading log collection unit 150 in a general terminal device having a function of connecting to the network 300 such as a personal computer, a tablet terminal device, and a smartphone. Can be configured by incorporating a dedicated program. Therefore, in practice, the terminal device 100 can be provided to a large number of users by downloading the dedicated program to a terminal device used by a large number of users via the network 300. Therefore, in the following description, an embodiment in which a large number of terminal devices 100 having the configuration shown in FIG. 1 are prepared will be described.

  On the other hand, as shown in the lower part of FIG. 1, the server device 200 includes a distribution target data storage unit 210, a reading log storage unit 220, a reading log extraction unit 230, an interest level distribution generation unit 240, a high interest level region extraction unit 250, and a trial reading. A data creation unit 260 is provided.

  The distribution target data storage unit 210 stores electronic book data to be distributed to each terminal device 100 and performs a process of distributing it to each terminal device 100 as necessary. The figure shows a state where electronic book data files F1, F2, and F3 are stored. The electronic book data files F1, F2, and F3 stored in the electronic book data storage unit 110 on the terminal device 100 side are files received from the distribution target data storage unit 210 on the server device 200 side.

  Of course, in practice, a large amount of electronic book data files are stored in the distribution target data storage unit 210, and the user uses the terminal device 100 that uses the desired electronic book data file as necessary. You can download it and use it. The electronic book data files F1, F2, and F3 downloaded in this way are not trial reading data but data constituting a complete electronic book. Therefore, the user can freely browse these downloaded electronic books, and as described above, the browsing history is collected as a reading log and transmitted to the server device 200.

  The reading log storage unit 220 is a component that stores a reading log transmitted from the reading log collection unit 150 on the terminal device 100 side via the network 300. As described above, in the case of the embodiment shown here, since reading logs of a large number of users are transmitted from a large number of terminal devices 100 connected to the network 300, the reading log storage unit 220 is transmitted from each terminal device. A process of sequentially storing the reading logs of a large number of transmitted users is performed.

  However, since such a reading log is personal information indicating a user's reading pattern and preference, if it is collected without the user's consent, there is a possibility that a legal or ethical problem may occur. Therefore, in order to actually collect reading logs from a large number of users using the present invention, it will be necessary to obtain the consent of the user to collect the reading logs in advance and ask for cooperation.

  For example, the distribution target data storage unit 210 is provided with a function to sell electronic books online, and an electronic book data file is distributed to a user who wishes to purchase an electronic book on a charge basis. At this point, you may ask to ask if you would like to cooperate in collecting reading logs. In practice, for example, users who cooperate with collection will be able to increase the number of collaborators by taking measures such as discount sales of electronic books. Alternatively, it is possible to operate a system that recruits monitors who cooperate in collecting reading logs on the condition that electronic books are distributed free of charge or at a low price, and has the users who have applied for cooperation cooperate.

  The server device 200 analyzes a large number of reading logs collected from such collaborator users, and automatically extracts test reading data from the original electronic book data by extracting portions suitable for the test reading. Processing will be performed. The sample reading data created in this way is distributed to other general users, and plays a role of stimulating purchase intention for these general users. Of course, a general user who purchased an electronic book can now cooperate with collecting a reading log as a monitor.

  In order to create sample reading data based on the collected reading log, the server apparatus 200 performs the following processing. First, the reading log extraction unit 230 extracts a reading log related to a creation target book, which is a creation target of trial reading data, from the reading log storage unit 220. For example, when the electronic book F1 (electronic book provided as the electronic book data file F1) is a creation target book for which trial reading data is to be created, the reading log extraction unit 230 is collected in the reading log storage unit 220. A reading log related to the electronic book F1 is extracted from reading logs of a large number of users.

  Subsequently, the interest level distribution creation unit 240 analyzes the extracted reading logs of a large number of users, obtains the interest level of the user for each individual part of the creation target book F1, and calculates the interest level for each part. Create an interest distribution showing the distribution of. For example, if “1 page” of a book is handled as “1 part”, the reading level is analyzed to obtain the degree of interest for each page of the book F1 to be created. ~ The distribution of the interest level up to the last page can be obtained. Here, the “degree of interest” is a parameter indicating the degree of interest of the user for each individual part (individual page) as a numerical value, and is an index indicating how “interesting” the user feels during browsing. .

  Of course, whether or not it feels “interesting” depends on individual differences among users. Therefore, in accordance with the principle of the present invention, it is possible to create sample reading data using the reading log of only one user. However, practically, it cooperates with a plurality of users, more preferably as many users as possible. It is preferable that the degree of interest is calculated assuming a virtual user having a standard preference. In addition, since the material for determining whether or not each part of the book felt “interesting” is only the reading log collected from the user, the degree of interest required here is an estimated value based on situation evidence only. Yes, and not directly related to the user's own evaluation. A specific method of defining this “interest level” will be described in detail in §3 and §4 below.

  Now, since “degree of interest” is a numerical parameter, the interest level distribution can be grasped as a change graph of numerical values from the first page to the last page of the book. The high interest level region extraction unit 250 performs a process of extracting a high interest level region where the interest level is equal to or higher than a predetermined threshold based on the interest level distribution. For example, when “1 page” of a book is set to “1 part”, one interest level is defined for each page, so the interest level distribution of a book of all 100 pages is composed of a list of 100 numerical values. Is done. Therefore, if only pages where the numerical value indicating the degree of interest is equal to or greater than a predetermined threshold value are extracted, page groups such as pages 15 to 20, 32 pages, 52 to 65 pages, and 92 to 100 pages are extracted. Will be. Each of these page groups is a high interest level region, and is a portion (ie, “interesting portion”) in which the user exhibits an interest level equal to or greater than a threshold value in the entire book.

  The principle of creating sample reading data in the present invention is that such an “interesting part” is divided in the middle, divided into a first half part and a second half part, and only the first half part is adopted as the sample reading data, and the second half part is intentionally dropped from the sample reading data. The idea is to make it happen. If only the first half of the high interest area that many users have evaluated as an “interesting part” is presented as trial reading data, the user who tried this first half will be urged to read the second half of the subsequent part. It will be. For this reason, it is possible to create trial reading data effective for inviting the user to purchase.

  In order to create sample reading data based on such a principle, the sample reading data creation unit 260 uses the electronic book data of the creation target book F1 stored in the distribution target data storage unit 210 and uses at least one high-ready data. A process of creating sample reading data including the first half of the interest level region but not the second half is performed. A specific method for creating the sample reading data will be described in detail in §6.

  Thus, the server device 200 can create trial reading data effective for inviting users to purchase by automatic processing based on reading logs of a large number of users. The process of creating the sample reading data does not require manual work, and the sample reading data can be efficiently created even when a large number of books are distributed as electronic books on the Internet.

<<< §2. Example of instruction input and reading log >>>
Next, a specific example of an instruction input taken by the instruction input unit 140 shown in FIG. 1 and a reading log collected by the reading log collection unit 150 will be described. As described above, the instruction input unit 140 specifies, from the user, an instruction to select a specific electronic book as a browsing target book or an instruction to cancel the selection, and a specific part of the selected electronic book as a browsing target part. An instruction input including an instruction is received, and based on the instruction input, the browsing control unit 120 reads data related to the browsing target part of the browsing target book from the electronic book data storage unit 110, and the display 130 based on the read data. A process of displaying the contents of the electronic book is performed on the top.

  In the case of the example shown in FIG. 1, since three electronic book data files F1, F2, and F3 are stored in the electronic book data storage unit 110, the user first browses one of these files. An instruction to select as a book is given, and then an instruction to specify a browsing target part in the selected browsing target book. The browsing target part is a part displayed on the screen of the display 130. Here, an example in which “one page” is handled as “one part” will be described. In other words, in this embodiment, one page is displayed on the screen of the display 130.

  FIG. 2 is a diagram illustrating an example of an electronic book browsing state in the terminal device 100 illustrated in FIG. 1. In this example, the i-th electronic book Fi (Ryunosuke Sasakawa “Kashiwa no Ito”) stored in the electronic book data storage unit 110 is selected as a browsing target book, and the j-th page Pj is browsed. It is an example when it is specified as a target part. As shown on the left side of the figure, in this example, the electronic book data is composed of a collection of data in page units, and the contents of the hatched page Pj are displayed on the display 130 shown on the right side of the figure. Is displayed on the screen.

  FIG. 3 is a diagram illustrating an example of user instruction input captured by the instruction input unit 140 illustrated in FIG. 1. In this example, as shown in the figure, 4 "selection instruction for browsing target book", "selection instruction for browsing target book", "instruction for specifying browsing target part (page)", and "instruction for adding auxiliary information" Classification instruction input is set.

  As the “browsing target book selection instruction”, an “open file” command and a “window selection” command are prepared. The “open file” command is input by, for example, clicking or tapping an icon to be viewed in a state where icons indicating the three electronic book data files F1, F2, and F3 are displayed. be able to. Alternatively, it can be input by an operation of selecting an “open file” command from the menu screen. The display state shown in FIG. 2 is a result of clicking the icon of the i-th electronic book data file Fi and selecting the electronic book Fi as a reading target book.

  The “select window” command is a command used when the electronic book viewing application program has a specification that allows a plurality of files to be opened simultaneously. Generally, in the case of an electronic book browsing application program incorporated in a personal computer, such a specification is often adopted. In this case, for example, when the user executes an “open file” command for both the files F1 and F2, two different windows are displayed on the screen of the display 130, and the contents of the file F1 are displayed in one window. Is displayed, and the contents of the file F2 are displayed in the other window.

  The “select window” command is a command for selecting one window as an active window when a plurality of windows are displayed as described above. For example, an operation of clicking one point in a window to be selected Can be entered. Normally, the activated window is displayed in the foreground, and the electronic book displayed in the active window is selected as the book to be browsed.

  On the other hand, “browsing target book selection cancellation instruction” is an instruction input for canceling the selection based on the “browsing target book selection instruction”. In this embodiment, as shown in FIG. , “Select another window” command, “End application” command, “Sleep” command, “Power OFF” command are prepared.

  Here, the "Close File" command is literally a command to close the currently open file. Click the "Close File" button or select the "Close File" command from the menu screen. Can be input. When the file is closed, the selection of the reading target book is canceled for the electronic book related to the file, and it is no longer the reading target book.

  In addition, as described above, the “select another window” command is a window that is different from the currently active window (the display window of the current reading book) when a plurality of windows are displayed. Is actually a command given by the same operation as the “window selection” command described above. For example, when two files F1 and F2 are opened at the same time and the window displaying the contents of the file F1 is active, one point in the window displaying the contents in the file F2 is selected. When a click operation is performed, the click operation becomes a “window selection” command for the file F2 and an instruction input for selecting the electronic book F2 as a new book to be browsed. The “select” command becomes an instruction input for canceling the selection of the electronic book F1.

  If the electronic book viewing application program has a specification that does not permit the simultaneous opening of a plurality of files, that is, if the specification allows only one file to be opened at all times, the file F1 is In the opened state, the “open file” command for the file F2 performs the same function as the “close file” command for the file F1.

  Further, the “application end” command, the “sleep” command, and the “power OFF” command shown in FIG. 3 are not originally commands that are input with the intention of canceling the selection of the reading target book, but in the embodiment shown here. In this case, it is handled as conforming to the “instruction to cancel the selection of the reading target book”.

  In other words, the “end application” command is a command for ending the electronic book browsing application program that is currently running, and if the command is input, the file that has been opened for use by the application program is automatically selected. And the same process as when the “close file” command is given. On the other hand, the “sleep” command and the “power off” command are commands for setting the terminal device 100 itself to a standby state or a stop state.

  Depending on the terminal device, when the “sleep” command is given, the application program does not end, the open file is not closed and the standby state is entered, and then the “wakeup” command is given. In some cases, the process to return to the state is performed, but in the case of the embodiment shown here, the “sleep” command is treated as the “instruction to cancel the selection of the book to be viewed”, and the original state is restored by the “sleep release” command. At that time, it is handled again as being given a “browsing target book selection instruction”.

  In addition, “instruction for designating a part (page) to be browsed” is an instruction input for designating a part (page in this embodiment) which is to be displayed on the display among the currently selected books to be browsed. In this embodiment, as shown in the figure, designation of “default page”, “next page” / “previous page”, and “specific page” are set.

  Here, the designation of “default page” is not an instruction input that is actually captured by the user's intentional page designation operation, but an instruction that is automatically performed in accordance with the instruction input of the “open file” command. Input. That is, when an input operation of the “open file” command is performed, as described above, the specified file is opened. At this time, it is determined whether or not the file is the first file to be opened. . In the case of a file opened for the first time, the “first page” is set as the “default page”, and an instruction input for automatically specifying the “first page” is made. On the other hand, in the case of a file that has been opened in the past, the “page to be viewed when the previous file was closed” is set as the “default page”, and the “page to be read when the file is closed” is automatically specified. The instruction input is made.

  After all, when the user opens a specific file with this "default page" specification function, if the file is the first file to be opened, an instruction is input to specify "first page" as the page to be viewed, otherwise In this case, an instruction is input to designate “the page that was displayed when the file was closed last time” as the browsing target page.

  On the other hand, the designation of “next page” and “previous page” and the designation of “specific page” are instruction inputs that are taken in by the user's intentional page designation operation. That is, the designation of “next page” is an instruction input performed by an operation such as the user clicking the “next page” button, and is next to the page currently displayed on the display (the current browsing target page). This is an input for designating the page as a new browsing target page. Similarly, the designation of “previous page” is an instruction input performed by an operation such as the user clicking the “previous page” button, and is in front of the page currently displayed on the display (the current browsing target page). This page is input for designating a new page to be browsed. The designation of “specific page” is an instruction input performed by an operation such as a user inputting a number, and is an input for designating an arbitrary page identified by a number or the like as a new browsing target page.

  The “instruction to add auxiliary information” shown at the end is an instruction input for the user to intentionally add some auxiliary information to a specific part of the book currently displayed on the display. . In the example shown in the figure, an example of designation input of “highlight” designation, “underline” designation, and “memo” attachment is shown. “Highlight” is specified to highlight a specific location on the display. For example, after clicking “Highlight Tool”, a specific display location on the display screen is dragged. The auxiliary information indicating that the specific display portion should be “highlighted” is added. Similarly, “underline” designation is an instruction to add an underline to a specific location on the display, and “memo” attachment is a memo (a character string entered by the user) at a specific location on the display. It is an instruction input to attach. Such a “function for adding auxiliary information” is a known function provided in a general electronic book browsing application program, and thus detailed description thereof is omitted here.

  As described above, the instruction input unit 140 of the terminal device 100 has a function of receiving various instruction inputs as shown in FIG. Here, the “browsing target book selection instruction” and the “browsing target book selection cancellation instruction” indicate which of the electronic books stored in the electronic book data storage unit 110 is to be selected as the browsing target book. The instruction input is designated, and the “instruction for designating the browsing target part” is an instruction input for designating a specific part of the selected browsing target book as the browsing target part. In the embodiment shown here, as described above, an instruction input for designating the browsing target part on the page is made, and the browsing control unit 120 displays the content of the page specified as the browsing target part on the display 130. Become.

  On the other hand, the reading log collection unit 150 collects the display history of the designated page as a reading log, and the interest level distribution creation unit 240 obtains the interest level for each page, thereby calculating the distribution of the interest level in units of pages. The high interest level region extraction unit 250 extracts a high interest level region as a region having a page as a minimum unit.

  Of course, depending on the terminal device, there is a case where “2 pages” is handled as “1 part” and 2 pages are displayed on the display screen. However, here, “1 page” is handled as “1 part”. A specific example will be described below.

  FIG. 4 is a diagram illustrating an example of a reading log collected by the reading log collection unit 150 illustrated in FIG. 1. In this example, for a specific electronic book, information that records the designation history of the browsing target part together with the time from when the instruction to select the reading target book is given until the instruction to cancel the selection is given. Collected as a unit reading log. Here, the unit reading log is data having a format of F [(P, T), (P, T), (P, T),..., (P, T)], and F is a browsing target. The file name of the electronic book selected as the book, P is the designated browsing target part (in this example, the designated page), and T is the time when the browsing target part is designated (for example, year, month, day, hour, minute, second) Data). The browsing target part P and the time T are recorded in a pair data format (P, T), and the browsing target part designation history from when the file F is selected until the selection is canceled. Are recorded as an enumeration of (P, T).

  The table shown in FIG. 4 shows an example of the reading log collected in this way. Here, L1, L2, and L3 are unit reading logs, P1 to P36 indicate pages 1 to 36 of each book, and T1 to T25 indicate log collection times. These unit reading logs are created when the following instruction input is given from the user.

  First, it is assumed that the user performs an operation of clicking the icon of the file F3 and inputs a “open file” command at time T1 in order to view the electronic book F3 for the first time. With this command, the file F3 is opened, and the electronic book F3 is selected as a reading target book. As a result, a container for the unit reading log L1 of F3 [] is created. Since the file F3 is a file opened for the first time, the “default page” is automatically designated in accordance with the input of the “open file” command, and the “first page”, that is, “P1” is the browsing target part. Specified as As a result, pair data (P1, T1) is created. The unit reading log L1 at this time is F3 [(P1, T1)].

  In this way, the user views P1 (first page) of the electronic book F3 displayed on the display screen, but finishes reading P1 and inputs an instruction to designate “next page” at time T2 that follows. As a result, the browsing target part is P2, and P2 (second page) is displayed on the display screen. At this time, pair data (P2, T2) is created, and the unit reading log L1 at this time is F3 [(P1, T1), (P2, T2)].

  The unit reading log L1 shown in the table shown in FIG. 4 is collected based on such user instruction input. Specifically, the user who has read P2 designates “next page” one after another, proceeds to P3, P4, and P5, designates “previous page” here, returns to P4, and then continues to “ When the “next page” is designated one after another, reading proceeds to P5, P6, and P7, and finally the “close file” command is input at time T10, the unit reading log L1 shown in the drawing is collected. . The last (φ, T10) is a log indicating that a “close file” command is given at time T10, and “φ” is a symbol indicating that no page is specified. By closing the file F3 at this point, the unit reading log L1 is completed.

  The unit reading log L2 collected subsequently is based on the following instruction input by the user. First, in order to view the electronic book F1, the user performs an operation of clicking the icon of the file F1 at time T11 and inputs a “open file” command. With this command, the file F1 is opened, and the electronic book F1 is selected as a reading target book. As a result, a container for the unit reading log L2 of F1 [] is created. Here, it is assumed that the file F1 is a file that has been opened in the past, and the previous “page to be viewed when closing the file” was P32. The fact that the last page to be browsed of this file F1 is P32 has a history in the previous unit reading log (not shown in FIG. 4) regarding the file F1, and as described above, the file F1. In response to the input of the “open file” command, “default page” is designated, and P32 is automatically designated as the page to be browsed. As a result, pair data (P32, T11) is created, and the unit reading log L2 at this time is F1 [(P32, T11)].

  Next, the user who has finished reading P32 designates “next page” one after another, proceeds to P33 and P34, designates “specific page” here, returns to P1, and then continues to “next page”. Designate and proceed to P2, specify "specific page" again and return to P34, then designate "next page", proceed to P35 and P36, and finally at time T19 When the “close file” command is input, the unit reading log L2 shown in the figure is collected. The last (φ, T19) is a log indicating that a “close file” command is given at time T19, and “φ” is a symbol indicating that no page is specified. By closing the file F1 in this way, the unit reading log L2 is completed.

  On the other hand, the last unit reading log L3 is based on the following instruction input by the user. First, in order to browse the electronic book F3 again, the user performs an operation of clicking the icon of the file F3 at time T20, and inputs an “open file” command. With this command, the file F3 is opened, and the electronic book F3 is selected as a reading target book. As a result, a container for the unit reading log L3 of F3 [] is created. Here, the file F3 is a file that was opened last time, and by referring to the unit reading log L1, it can be recognized that the previous “page to be viewed when closing the file” was P7 ((P7 of L1 , T9)), P7 of the file F3 is displayed on the display screen. That is, “default page” is designated in accordance with the instruction input of the “open file” command for the file F3, and P7 is automatically designated as the browsing target page. As a result, pair data (P7, T20) is created, and the unit reading log L3 at this time is F3 [(P7, T20)].

  Next, when the user who has read P7 designates “next page” one after another, proceeds to P8 to P11, and finally enters the “close file” command at time T25, the unit reading shown in the figure. The log L3 is collected. The last (φ, T25) is a log indicating that a “close file” command is given at time T25, and “φ” is a symbol indicating that no page is specified. By closing the file F3, the unit reading log L3 is completed.

  The unit reading log illustrated in FIG. 4 is collected based on user instruction input as in the above example. Note that the instruction input in the above example is shown as an example, and a similar reading log can be collected by another user operation. For example, in the above instruction input, an example in which the currently selected file is deselected by giving a “close file” command has been shown, but as described above, the “select another window” command, Even when an “application end” command, a “sleep” command, a “power off” command, or the like is performed, the selection of the reading target book is canceled and each unit reading log is completed.

  In addition, although description about the format of the reading log when the instruction | indication which adds auxiliary information is given is abbreviate | omitted here, the data which show the position which adds auxiliary information, and the information which shows auxiliary information itself were included As long as it is data, the reading log may be composed of any format of data.

<<< §3. Example of negative interest definition >>>
Next, a specific example of interest level definition obtained by the interest level distribution creating unit 240 shown in FIG. 1 will be described. As already described, the “degree of interest” in the present invention is a parameter indicating the degree of interest of the user for each individual part of a book, and the greater the value of “degree of interest”, the more users The part is evaluated as “interesting”. In the case of the example shown here, since “1 page” of the book is handled as “1 part”, an evaluation value of “interesting”, that is, “interesting” is given to each page. become.

  Here, for a specific book, a clue for determining the degree of interest for each page is a reading log of many users regarding the specific book transmitted from the terminal device. In the case of the example shown in FIG. 4, the reading log is information indicating when and which page of which book is displayed on the display. Of course, it cannot be distinguished whether the user is actually reading or whether the corresponding page is simply displayed on the display screen. Therefore, in the present invention, when a specific page (part) is displayed on the display screen, it is assumed that the user has browsed the page (part), and the processing described below is executed.

  Now, there are several variations in the method of estimating the degree of interest of each user page based on the reading log as shown in FIG. Here, first, a sign is assigned to the interest level, and a positive interest level and a negative interest level are defined. The positive degree of interest is a value that expresses the degree of “interesting” with a numerical value. The larger the numerical value, the more the user feels “interesting”. Conversely, the negative interest level is a value expressing the degree of “not interesting” with a numerical value, and the larger the numerical value, the more the user feels “not interesting”. In §3, first, a definition example of the negative interest level will be described.

  Here, the parameter used as the negative interest level focuses on “interruption of reading”. Usually, the user interrupts reading at several points in the middle before reading a book. In addition, the user tends to read the portion that is considered “interesting” at once, and to stop reading at the portion where the “interesting” has dropped. Therefore, by analyzing the reading log and recognizing the place where reading was interrupted and giving a negative interest level to the interrupting place, the interest level of the place where many users are likely to interrupt reading. Will decline. In other words, for each page, the degree of interest is defined by the deduction method.

  As described in §2, the instruction input unit 140 receives an instruction to cancel the selection of the electronic book (that is, an instruction to interrupt the browsing of the book) when a command to close the data file of the electronic book is given. It can be recognized as given. Then, the reading log collection unit 150 can collect the browsing target part at the time when the instruction is given as a reading log indicating the browsing interruption part for the electronic book. Therefore, the interest level distribution creating unit 240 can grasp the browsing interruption part of each electronic book based on the reading log.

  Of course, as illustrated as “browsing target book selection cancellation instruction” in FIG. 3, in addition to the command to close the data file, a command handled as an instruction to cancel the selection of the electronic book can be set. That is, the instruction input unit 140 gives an instruction to set the “terminal device itself” or “the electronic book browsing function of the terminal device (for example, an electronic book browsing application program)” to a standby state or a stopped state while browsing an electronic book. When it is received, it can be recognized that an instruction to cancel the selection of the electronic book is given. Also in this case, the reading log collecting unit 150 can collect the browsing target part at the time when the instruction is given as a reading log indicating the browsing interruption part for the electronic book, and the degree-of-interest distribution creating unit 240 Based on this reading log, the browsing interruption part of each electronic book can be grasped.

  FIG. 5 is a diagram illustrating a browsing interruption part that the interest level distribution creating unit 240 illustrated in FIG. 1 recognizes based on the reading log illustrated in FIG. 4. As described in §2, when a user reads an electronic book using the terminal device 100, a reading log as illustrated in FIG. 4 is collected by the reading log collecting unit 150, and the server device is connected via the network 300. 200 is transmitted to the reading log storage unit 220. The reading log is extracted by the reading log extraction unit 230 for each electronic book, and is analyzed by the interest level distribution generation unit 240.

  For example, if the unit reading log L1 is analyzed, the electronic book F3 can recognize that browsing is interrupted with P7 as the last page to be browsed, and if the unit reading log L2 is analyzed, the electronic book F1 displays P36. It can be recognized that browsing has been interrupted as the last browsing target page, and by analyzing the unit reading log L3, the electronic book F1 can recognize that browsing has been interrupted with P11 as the last browsing target page. Therefore, from the unit reading logs L1, L2, and L3, P7 and P11 are recognized as the browsing interruption part for the electronic book F3, and P36 is recognized as the browsing interruption part for the electronic book F1.

  In this way, the interest level distribution creation unit 240 analyzes the reading logs of a large number of users, recognizes the browsing target part at the time of deselecting the electronic book as the browsing interruption part of the electronic book, If the cumulative number of interruptions indicating the number of times the part has become a browsing interruption part is obtained, the cumulative distribution number of interruptions can be counted as a negative interest degree to create an interest degree distribution.

  For example, in the case of the above example, for the electronic book F3, based on the unit reading logs L1 and L3, a process of increasing the count value of the cumulative number of interruptions by 1 is executed for P7 and P11. If such a count process is performed on reading logs obtained from a large number of users, an interest distribution as shown in FIG. 6 can be obtained. The graph shown in FIG. 6 is a graph showing the interest level distribution for a certain electronic book. Here, the horizontal axis of the graph is the page of the electronic book, the left end indicates the first page, and the right end indicates the last page. The vertical axis of the graph represents the cumulative number of interruptions for each page. This distribution of interest is an aggregate value that counts the number of times that each page has become a browsing interruption part based on reading logs from a large number of users. Although it is a discrete value in page units, continuous graphs are shown in the figure for convenience of explanation.

  Note that even when the same page continues to be displayed on the display screen for an excessively long time, it can be handled as if browsing has been interrupted. For example, if the same page continues to be displayed for 5 minutes or more even though the average browsing time for one page is about 1 minute, the user is not in a normal browsing state. Probability is high. If the user is performing normal reading, it is common to input an instruction such as specifying “next page” and turn the page after a time of about one minute has passed. Nevertheless, in the case where no operation has been performed for 5 minutes, the user is not actually reading, has been left for some reason, or is doing another task. Probability is high.

  In the case of the embodiment shown here, the reading log collecting unit 150 has a function of collecting, as illustrated in FIG. 4, the time T when the instruction is given together with an instruction for specifying the browsing target part P as a reading log. ing. Therefore, the interest level distribution generation unit 240 recognizes from the reading log the display time during which the same browsing target part (the same page in this example) is continuously displayed, and the display time is a predetermined time. If the allowable range (for example, 5 minutes) is exceeded, the browsing target part may be handled as a browsing interruption part.

  For example, in the unit reading log L1 shown in FIG. 4, when the time T7 and the time T8 are 5 minutes or longer, the user gives an instruction to display the page P5 at the time T7, and then actually reads the book. Since it can be estimated that it has been interrupted, the process of counting P5 as a browsing interrupt part may be performed.

  In the graph shown in FIG. 6 obtained in this way, a page having a large number of interruption accumulation times is a page where many users have interrupted reading. Therefore, as a general user's tendency to read, if it is assumed that there is a tendency that the part that is thought to be “interesting” is read at a stroke and the reading is interrupted at the part where “interesting” has broken down, A page with a large cumulative count value can be said to be “not interesting”. Therefore, if this cumulative number of interruptions is treated as a parameter indicating a negative degree of interest, as shown on the left side of the graph, the degree of interest decreases as the cumulative number of interruptions increases, and the interest decreases when the cumulative number of interruptions decreases. The degree will be higher.

  In the embodiment shown in FIG. 6, based on this interest level distribution, a portion where the interest level is equal to or greater than a predetermined threshold value Th (the interrupt accumulation count is a negative interest level. Therefore, in the graph of FIG. The portion where the number of times is equal to or less than the horizontal line Th) is extracted as the high interest level regions H1, H2, and H3. Note that the region H0 is also a portion where the cumulative number of interruptions is equal to or less than the horizontal line Th. However, in this embodiment, since the first region of the electronic book is not extracted as a high interest region, it is illustrated. The area H0 is not a high interest area. The reason why such an operation is used is that, in general, the top area of a book tends to be read without being interrupted even if it feels "not interesting".

  Eventually, in the example shown in FIG. 6, three high interest areas H1, H2, and H3 are extracted. These areas are all composed of a group of continuous pages, and many users read at a stretch without interruption during reading, that is, many users feel "interesting". This is the part that can be estimated to be. Of course, the width and total number of the high interest level regions vary depending on the value of the threshold Th, and depending on the setting of the threshold Th, the high interest level region whose width is wider as it extends over almost all pages of the book. May be extracted, or no high interest area may be extracted. Therefore, for practical purposes, it is preferable to set the threshold Th to various values for some books and determine the threshold Th from which an appropriate region of high interest is extracted by trial and error.

<<< §4. Example of positive interest definition >>>
In §3 described above, an example in which the cumulative number of interruptions is used as a negative degree of interest has been described. This can be said to be a technique of recognizing a point of interest by recognizing a point where many users interrupt reading as a “not interesting” point. Here, conversely, some examples using positive interest will be described. This can be said to be a method of positively adding points where many users may find “interesting”.

<§4-1. Cumulative display count>
A first example of a parameter that can be used as a positive degree of interest is the cumulative number of displays. For example, a book of 100 pages is read by 50 users, the reading log from each user is analyzed, and the number of times each page is displayed on the display screen (that is, designated as a browsing target part). Let's try to accumulate the number of times). In this case, if each user displays each page once and browses it, it is expected that a cumulative value of almost 50 times will be obtained for any page. However, for example, when viewing the unit reading log L1 in FIG. 4, P4 and P5 are displayed twice. This is because the user designates “Previous Page” during browsing and rereads P4 and P5. In the unit reading log L2, an operation of returning from P34 to P1 and P2 is performed, and pages that have been browsed in the past are reread.

  Thus, even when a user browses a 100-page book, the number of times each page is displayed is not necessarily equal, and depending on the page, it may be repeatedly viewed twice or three times. It is expected that the page that is repeatedly browsed in this way includes some content that attracts the user's interest. Therefore, a page with a large cumulative display count for a large number of users is a page that many users have read back many times, and it is highly likely that many users feel “interesting”. From such a viewpoint, this display cumulative number can be used as a positive interest level.

  When the display cumulative number is used as the positive interest level, the interest level distribution creating unit 240 counts the display cumulative number of individual parts (pages) of the book for which the trial reading data is generated as the positive interest level. To create an interest distribution. Specifically, an operation of counting the number of times each page appears from the unit reading log as shown in FIG. 4 may be performed. In addition, when reading is interrupted and restarted later, when the terminal device adopts a specification that automatically redisplays the page at the time of interruption as the “default page” at the time of restarting, Pages are counted redundantly when suspended and resumed. However, as stated in §3, the page where reading was interrupted should be treated as a page with rather low interest.

  For example, in the example shown in FIG. 4, in the unit reading log L1, since reading is interrupted at P7, and reading is resumed at P7 in the unit reading log L, the count value of the number of display times of P7 is 2. Times. However, since P7 is a portion where reading is interrupted, it is not preferable to count twice as many as the cumulative display count is treated as a positive interest level. Therefore, in practice, for example, it is preferable to adopt an operation in which, for example, a display page at the time when an instruction to cancel selection as a reading target book is given is excluded from the cumulative number counting target. In the case of the above example, (P7, T9) of the unit reading log L1 is excluded from the target of counting the cumulative number.

<§4-2. Integrated value of display time>
A second example of a parameter that can be used as a positive interest level is an integrated value of display time. This is the degree of interest defined based on the hypothesis that “the part read over time is an interesting part”. In other words, since the user will read the “interesting” portion over time, if the integrated value of the display time is obtained for each page, the integrated value can be used as a positive degree of interest. This is based on the idea of

  When the integrated value of the display time is used as a positive interest level, the reading log collecting unit 150 collects the time at which the instruction is given as an reading log together with an instruction to specify the browsing target part (page). The degree distribution creation unit 240 recognizes the display time of each browsing target part from this reading log, and counts the integrated value of the display time for each part of the book for which trial reading data is created as a positive interest degree. What is necessary is just to create an interest degree distribution. For example, if a reading log having a data format as shown in FIG. 4 is created, the display time of each page can be calculated by obtaining the time difference at the time when the browsing target page is switched. For example, by referring to the pair data (P1, T1) and (P2, T2) at the beginning of the unit reading log L1, the display time of P1 can be calculated by the calculation “T2-T1”.

  Actually, since the reading speed of each user varies considerably, even if a display time of “1 minute” is obtained for a certain page from the reading log, it takes a long time for a fast reading user. However, it is often the case that it is a short time for slow-reading users. Therefore, the fact that the display time of a certain page was “1 minute” alone cannot accurately grasp whether the user has read the page at a normal speed or carefully.

  Therefore, in order to obtain a higher degree of interest, the reading log storage unit 220 stores the reading log separately for each user, and the interest level distribution generation unit 240 displays the individual parts (pages). After normalizing the time for each user, the integrated value of the normalized display time may be counted as the positive interest level to create the interest level distribution.

  The unit reading logs L1, L2, and L3 shown in FIG. 4 do not have information indicating which user the reading log is, but in order to distinguish the reading log for each user, the user identification code Ui is assigned to each user in advance. And the user identification code may be added to the reading log. For example, if each unit reading log shown in FIG. 4 is created by the browsing action of the user U1, the file names of each unit reading log are L1 (U1), L2 (U1), L3 (U1), and the like. What is necessary is just to add user identification code U1. Since the interest level distribution creation unit 240 can classify the individual unit reading logs for each user based on the user identification code, the display time normalization process can be performed for each user.

  FIG. 7 is a graph illustrating an example of processing in which the interest level distribution creating unit 240 normalizes the display time for each user. Graphs U1, U2, and U3 shown in FIG. 7A are graphs showing the display time distribution of each page for three users U1, U2, and U3. As shown in the figure, the average display speed of three users (considered as reading speed) is different, and even for the same user, the display speed is different for each page. The difference in display speed for each page for the same user is a parameter indicating the degree of interest.

  Therefore, the minimum display time and the maximum display time per page are obtained for each user. For example, the minimum display time is set to 0, the maximum display time is set to 1, and all the display times are within the range of 0 to 1. Normalize like this. Graphs U1 ', U2', U3 'shown at the bottom of Fig. 7 (b) are obtained by performing such normalization on the graphs U1, U2, U3 shown in Fig. 7 (a). It is a graph. Of course, the normalization method is not limited to the above method. For example, for each graph, normalization may be performed so that the area under the graph becomes a constant value. It doesn't matter.

  The graph Σ shown in the upper part of FIG. 7 (b) is the sum of the normalized graphs U1 ′, U2 ′, U3 ′, and is an integrated value of the normalized display time for all users. Become. Thus, if the display time distribution of each user is normalized and then integrated, a more accurate interest level distribution in which the difference in reading speed for each user is corrected can be obtained.

  As described in §3, if the same page continues to be displayed on the display screen for an excessively long time, the user is likely not in the normal browsing state. For example, if the unit reading log calculates the display time of a specific page as 5 minutes, the user is not actually reading and has left the office for some reason, There is a high possibility of going. Therefore, the long display time of 5 minutes does not indicate that “the user is reading the page carefully with interest”, and it is not preferable to integrate this as a parameter indicating the degree of positive interest. Therefore, when the display time during which the same browsing target part (page) continues to be displayed exceeds a predetermined allowable range (for example, 5 minutes), the interest level distribution creation unit 240 accumulates the display time. It is preferable to take an operation that does not include the value.

<§4-3. Additional frequency of auxiliary information>
A third example of a parameter that can be used as a positive interest level is the frequency of adding auxiliary information. Here, the “auxiliary information” is any information that is intentionally added by the user to a specific part of the book currently displayed on the display as described in §2, for example, “highlight”. Information, such as designation of "underline", attachment of "memo". Of course, the auxiliary information is not limited to such information, and may be various marks other than “highlight” and other arbitrary information.

  In order for the terminal device 100 to have such an auxiliary information addition function, the instruction input unit 140 has a function of receiving an instruction input for adding auxiliary information to a specific location of the selected electronic book, and browsing control. The unit 120 may store the auxiliary information and have a function of adding and displaying the stored auxiliary information when the specific portion is displayed on the display 130. In this case, the reading log collection unit 150 has a function of collecting a reading log including an instruction to add the auxiliary information and transmitting the reading log to the server device 200 via the network 300. Then, the interest level distribution creating unit 240 creates the interest level distribution by counting the addition frequency of the auxiliary information for each part (page) of the book for which the trial reading data is created as a positive interest level. Can do.

  An arbitrary method can be set as the additional frequency counting method. For example, regardless of the type and content of the added auxiliary information, when some auxiliary information is added to a specific part of the book, the frequency is counted as 1, and the total number of places where the auxiliary information is added is counted as the additional frequency. Can be taken. Or you may set the frequency value according to the amount of the auxiliary information added. For example, when “highlight” designation or “underline” designation is added as auxiliary information to a character string, the number of characters in the character string to be designated may be counted as a value indicating the frequency. Further, when “memo” is attached as auxiliary information to a specific portion of the book, the number of characters in the memo may be counted as a value indicating the frequency.

  It can be estimated that such a page with a high frequency of supplemental information is that many users are interested in the contents of the page. Therefore, the auxiliary information addition frequency can be used as a parameter indicating a positive interest level.

<§4-4. Extraction of high interest area>
The graph shown in FIG. 8 is a graph showing a positive interest distribution for one electronic book. Here, the horizontal axis of the graph is the page of the electronic book, the left end indicates the first page, and the right end indicates the last page. Moreover, what is necessary is just to take any of the positive interest level described so far, ie, the display frequency | count, the integrated value of display time, and the additional frequency of auxiliary information, as a vertical axis | shaft of the graph.

  In the graph (interesting degree distribution) shown in FIG. 8 obtained in this way, a page with a high degree of interest shown on the vertical axis is a page in which many users are interested. Therefore, the high interest level region extraction unit 250 may extract a region where the interest level is equal to or higher than the predetermined threshold Th based on the interest level distribution as the high interest level region. In the example shown in FIG. 8, high interest level regions H1, H2, and H3 are extracted as regions where the interest level is equal to or higher than the threshold Th. All of these areas are composed of a group of consecutive pages, and many users have shown some interest, that is, many users can estimate that they will find it interesting. It turns out that.

  Of course, also in this case, the horizontal width and the total number of the high interest level regions vary depending on the value of the threshold Th, and depending on the setting of the threshold Th, Only one area may be extracted, or no high interest area may be extracted. Therefore, for practical purposes, it is preferable to set the threshold Th to various values for some books and determine the threshold Th from which an appropriate region of high interest is extracted by trial and error.

  Heretofore, as the positive interest level, three parameters such as the display cumulative number, the display time integrated value, and the auxiliary information addition frequency are exemplified, and any one of these is shown as the vertical axis of the graph shown in FIG. Although the example to employ | adopt was described, it is also possible to obtain | require an interest degree distribution using the interest degree which combined these three parameters. In that case, it is preferable that each parameter is weighted and added. That is, a value indicating the degree of interest may be defined by an expression such as interest degree = k1 · (accumulated number of displays) + k2 · (integrated value of display time) + k3 · (additional frequency of auxiliary information). Here, k1, k2, and k3 are weighting coefficients for each parameter.

  If necessary, the negative interest level described in §3 and the positive interest level described in §4 can be added in consideration of the sign. That is, the value indicating the degree of interest is expressed by an expression such as interest degree = k1 · (display cumulative number) + k2 · (display time integrated value) + k3 · (additional frequency of auxiliary information) −k4 · (interrupted cumulative number). Define it. Here, k4 is a weighting coefficient for the negative interest level “accumulated number of interruptions”.

  In any case, based on the obtained interest level distribution, an area where the interest level is equal to or higher than a predetermined threshold is extracted as a high interest level area.

<<< §5. Sample of sample reading data >>>
Here, a specific method for creating sample reading data by the sample reading data creating unit 260 will be described. As described above, the trial reading data creation unit 260 performs a process of creating trial reading data based on the electronic book data of the creation target book stored in the distribution target data storage unit 210. This process is a process in which a part of the original electronic book data is extracted and used as sample reading data, and the basic policy includes data in the first half of at least one high interest area, but in the latter half. This is to create sample reading data that does not include the data.

  Here, some specific examples of sample reading data based on the basic policy will be described with reference to FIG. FIG. 9 shows the contents of one electronic book with a common page axis in the horizontal direction. The bar in FIG. 9A shows the configuration of the original electronic book. That is, this electronic book is composed of four chapters of chapters 1 to 4. The left end of the bar is the first page of the electronic book, and the right end of the bar is the last page of the electronic book. FIG. 9B shows the high interest area extracted by the high interest area extraction unit 250. In the illustrated example, high interest areas H1, H2, and H3 are extracted, and these areas are estimated to be recognized by many users as “interesting” portions.

  Hereinafter, with reference to FIGS. 9 (c) to 9 (f), an example of a specific method for creating sample reading data by the sample reading data creation unit 260, that is, the original electronic data shown as a bar in FIG. 9 (a). An example of a method for creating a sample reading data by extracting a part of a book will be described.

  In any case, the division position is set at an arbitrary position in the high interest area. Dotted lines D1, D2, and D3 shown in the figure indicate division positions set for the high interest level regions H1, H2, and H3, respectively. The division position is a position at which the contents are cut off in the sample reading data. Theoretically, the division position may be set at an arbitrary position in each high interest area. However, since the effect of the present invention can be achieved by censoring the sample reading data at a position that causes the sample reading user to urge to read the continuation, the most suitable position for obtaining such an effect is divided into the divided positions. It is preferable to set to. From such a viewpoint, the inventor of the present application considers that the center position of the high interest level region is the optimum position as the division position.

  In the embodiment described so far, since “one page” of the book is handled as “one part”, the high interest area is an area having a page as a minimum unit, but the division position is However, it is not always necessary to separate the pages from each other, and the division position may be set at the middle position of one page, or the division position may be set at the middle position of one line. If the division position is set ignoring the paragraph of the sentence, it will be divided in the middle of the sentence, and the end of the sample reading data may be in the middle of the sentence, but it is only the sample reading data, so it will be interrupted in the middle of the sentence There is no problem even if it is.

<§5-1. Excerpt example 1>
The first extraction method is a method in which a division position is set at an arbitrary position in a certain high interest level area, and data from the beginning of the high interest area to the division position is used as sample reading data. Specifically, the sample reading data A1 shown in FIG. 9 (c) is an example of the sample reading data created by such an extraction method. That is, the high interest area H1 is divided at the division position D1, and data from the top of the high interest area H1 to the division position D1 is extracted as the trial reading data A1. In short, the sample data A1 is an excerpt of only the first half of the high interest area H1. Since the sample reading data ends in the middle of the high interest area H1 that many users find “interesting”, the user can be prompted to read the continuation. Can motivate.

  Since the beginning of the high interest level area H1 is not necessarily the beginning of the sentence, the sample reading data A1 obtained by this excerpt method may be in the middle of the sentence, both at the beginning and at the end. However, since only the first half of the high interest area, that is, only the first half of the part that many users felt “interesting”, it gave the test-reading user an “interesting” impression and stimulated the willingness to purchase. A sufficient effect is expected.

  FIG. 9 (c) shows an example in which the first half of the high interest area H1 is extracted and used as the sample reading data A1, but of course, the first half of the high interest areas H2 and H3 are extracted. The sample reading data may be used, or the data obtained by enumerating the first half of each high interest area may be used as the sample reading data.

<§5-2. Excerpt example 2>
The second excerpt method is a method in which a division position is set at an arbitrary position in a certain high interest level area, and data up to a position retroactive by a predetermined amount from this division position is used as sample reading data. Specifically, the sample reading data A2 shown in FIG. 9 (d) is an example of the sample reading data created by such an extraction method. That is, the high interest area H1 is divided at the division position D1, and data up to a position that is a predetermined amount m away from the division position D1 is extracted as the trial reading data A2.

  Here, the predetermined amount m may be specified in units of the number of pages, or may be specified in units of the number of lines or the number of characters. The merit of the extract example 2 is that the amount of the sample reading data to be created can be unified to m. For example, when there is a restriction such as “the sample reading data should be within 10 pages” or “the sample reading data should be within 5000 characters” due to a request from a publisher or the like, the method of this excerpt example 2 can be adopted. If so, it is possible to create sample reading data having a desired amount.

  In this excerpt 2 as well, since the sample reading data ends in the middle of the high interest area H1 that many users feel “interesting”, the sample reading user may be prompted to “read more”. It is possible to invite the user of the trial reading to purchase. Of course, data up to a position that is a predetermined amount m away from the division positions D2 and D3 of the high interest areas H2 and H3 may be extracted as sample reading data.

<§5-3. Excerpt example 3>
In the third excerpt method, a division position is set at an arbitrary position in a certain high interest area, and from this division position to a position that goes back to the beginning of the paragraph, section, chapter, or entire book including the division position. It is a method of using the data of as a trial reading data. Specifically, the sample reading data A3 shown in FIG. 9 (e) is an example of the sample reading data created by such an extraction method. That is, the high interest area H1 is divided at the division position D1, and the beginning of the chapter including the division position D1, that is, data from the beginning of chapter 2 to the division position D1 is extracted as the sample reading data A3. .

  The merit of the extract example 3 is that the beginning part of the sample reading data is located at a well-delimited position, so that there is no sense of incongruity when starting to read the sample reading data. Of course, even though there is no sense of incongruity at the beginning of the sample reading data, the end of the sample ends in the middle of the “interesting” part, so that the user who wants to read it can generate an urge to read more. It is possible to invite the purchase reading user's willingness to purchase.

  FIG. 9 (e) shows an example in which the sample reading data A3 is extracted with reference to the division position D1 of the high interest area H1, but of course, the division positions D2 and D3 of the high interest areas H2 and H3 are shown. The sample reading data may be extracted based on the above. The sample reading data extracted with reference to the division position D2 starts from the beginning of Chapter 3, and the sample reading data extracted with reference to the division position D3 starts from the beginning of Chapter 4.

  Further, the above example is an example in which the position that goes back to the beginning of the chapter is set as the beginning position of the sample reading data. Alternatively, data up to a position going back to the top of the entire book may be used as trial reading data.

<§5-4. Excerpt example 4>
The fourth extraction method is a method that is slightly different from the conventional extraction method. In other words, the conventional excerpt method is based on the concept of extracting a part of the original electronic book data shown in FIG. The excerpt method is based on the concept of deleting a part of the original electronic book data shown in FIG. As a result, a part of the original electronic book data is extracted, but the basic concept is slightly different.

  That is, in the fourth excerpt method, a division position is set at an arbitrary position of each high interest level area, and each high interest level area is divided into a first half portion and a second half portion respectively with the set division position as a boundary. Then, a procedure is adopted in which data obtained by excluding the latter half of each high interest area from the electronic book data of one book stored in the distribution target data storage unit 210 is used as trial reading data.

  Specifically, the sample reading data A4 shown in FIG. 9 (f) is an example of the sample reading data created by such an extraction method. The sample reading data A4 is composed of an aggregate of three partial data A4-1, A4-2, and A4-3, and includes a considerable content of the original electronic book. However, the latter half of each high interest area H1, H2, H3 is excluded. In other words, the second half of the “interesting” part is missing. Therefore, an urge to “read the missing part” can be generated for the sample reading user, and the purchase reading user can be invited to purchase.

<<< §6. How to create trial reading data >>>
So far, the electronic book trial reading data creation apparatus according to the present invention has been described. However, the present invention does not necessarily need to be executed using the above-described trial reading data creation apparatus. The basic technical idea of the present invention can also be grasped as a method of creating sample reading data for an electronic book provided using a network. In executing the creation method, the sample reading described above has not necessarily been performed. There is no need to use a data creation device.

  The electronic book trial reading data creation method according to the present invention includes the steps shown in FIG. First, step S1 is a reading log transmission stage executed by the terminal device. At this stage, based on the user's instruction input, the function of displaying the contents of the electronic book on the display and allowing the user to view it is displayed. The terminal device that has the user collects the browsing history of the user as a reading log, and transmits this via the network.

  Subsequent steps S2 to S5 are stages executed by the server device that creates the sample reading data. First, in step S2, a reading log reception stage is performed in which the server device receives reading logs transmitted from a plurality of terminal devices via a network. In step S3, the server device analyzes the received reading log to obtain an interest level indicating the degree of interest of the user for each part of the specific electronic book, and the interest level for the specific electronic book. An interest distribution creation stage for creating a distribution is performed. Furthermore, in step S4, a high interest level region extraction step is performed in which the server device sets a threshold value for the interest level of the interest level distribution and extracts a high interest level region where the interest level is equal to or higher than the threshold value. In the last step S5, the server apparatus extracts the first half of the data from the high interest area based on the electronic book data of the specific electronic book or deletes the second half of the data from the high interest area. A trial reading data creation stage for creating trial reading data is performed.

  In addition, in the interest level distribution creation stage in step S3, by analyzing the reading log, a browsing interruption part estimated to be interrupted by the user is recognized, and each part of the electronic book becomes a browsing interruption part. An interest distribution can be created by counting the cumulative number of interruptions as a negative interest. Alternatively, in the interest level distribution creation stage of step S3, by analyzing the reading log, the cumulative number of displays, the integrated value of the display time, or the frequency of adding auxiliary information by the user is counted as the positive interest level. It is also possible to create an interest distribution.

100: Terminal device 110: Electronic book data storage unit 120: Browsing control unit 130: Display 140: Instruction input unit 150: Reading log collection unit 200: Server device 210: Distribution target data storage unit 220: Reading log storage unit 230: Reading Log extraction unit 240: Interest level distribution creation unit 250: High interest level region extraction unit 260: Test reading data creation unit 300: Network (Internet)
A1 to A4: Test reading data A4-1 to A4-3: Each part of the test reading data D1 to D3: Division positions F1, F2, F3: Electronic book data file Fi: Data file H0 of the book to be viewed H0: First area of the electronic book H1 to H3: High interest area L1 to L3: Reading log m: Predetermined amount (predetermined number of pages / predetermined number of characters)
P1 to P36: Browsing target part (browsing target page)
Pj: Part to be browsed (page to be browsed)
S1 to S5: Steps T1 to T22 of the flowchart: Time Th: Threshold values U1 to U3 of interest level: Display time graphs U1 'to U3' of each user: Normalized display time graph φ of each user: Page Unspecified symbol Σ: Graph of the integrated value of normalized display time

Claims (21)

A device that creates trial reading data for an electronic book provided using a network,
A terminal device and a server device connected to the network;
The terminal device
An electronic book data storage unit for storing electronic book data distributed from the server device;
A display that displays the contents of the e-book;
An instruction to receive an instruction input including an instruction to select a specific electronic book as a browsing target book or an instruction to cancel the selection and an instruction to specify a specific part of the selected electronic book as a browsing target part. An input section;
Based on the instruction input, a reading control unit that reads data related to the browsing target part of the browsing target book from the electronic book data storage unit, and displays the contents of the electronic book on the display based on the read data;
A part or all history of the instruction input is collected as a reading log, and a reading log collecting unit that transmits this to the server device via the network;
Have
The server device
A distribution target data storage unit that stores electronic book data to be distributed, and distributes the electronic book data to the terminal device as necessary;
A reading log storage unit for storing a reading log transmitted from the reading log collection unit;
From the reading log storage unit, a reading log extraction unit that extracts a reading log related to a creation target book to be created as trial reading data;
By analyzing the extracted reading log, the interest level indicating the degree of interest of the user for each individual part is obtained for the creation target book, and the interest level distribution indicating the distribution of the interest level in parts is created. A degree distribution generator,
A high interest region extraction unit that extracts a high interest region where the interest is equal to or higher than a predetermined threshold based on the interest distribution;
Based on the electronic book data of the creation target book stored in the distribution target data storage unit, the trial reading data including at least one first half data but not including the second half data for the high interest area is created. A sample reading data creation unit,
An electronic book trial reading data creation device characterized by comprising: In the trial reading data creation device according to claim 1,
A plurality of terminal devices,
The reading log storage unit stores reading logs of a plurality of users transmitted from each terminal device,
The reading log extraction unit extracts reading logs of a plurality of users related to the creation target book,
An interest level distribution creating unit creates an interest level distribution by analyzing reading logs of a plurality of users, and an electronic book trial reading data creation device. In the sample reading data creation device according to claim 1 or 2,
The electronic book data is composed of a collection of page unit data,
The instruction input unit accepts an instruction input including an instruction to specify a browsing target part on a page,
The browsing control unit displays the contents of the specified page on the display,
The reading log collection unit collects the display history of the specified page as a reading log,
The interest level distribution creation unit creates the interest level distribution indicating the distribution of the interest level in units of pages by obtaining the interest level for each page,
A trial reading data creation device for an electronic book, wherein the high interest area extraction unit extracts a high interest area as an area having a page as a minimum unit. In the trial reading data creation device according to any one of claims 1 to 3,
The interest distribution creation unit recognizes the browsing target part at the time of deselecting the electronic book as the browsing suspended part of the electronic book, and for each part of the electronic book, shows the cumulative number of interruptions indicating the number of times it became the browsing suspended part. An electronic book trial-reading data creation device characterized in that an interest level distribution is created by counting the cumulative number of interruptions as a negative interest level. In the trial reading data creation device according to claim 4,
An electronic book trial reading data creation device, wherein the high interest area extraction unit does not extract the top area of an electronic book as the high interest area. In the sample reading data creation device according to claim 4 or 5,
When the instruction input unit is given an instruction to close the electronic book data file, the instruction input unit recognizes that the instruction to cancel the selection of the electronic book is given,
A reading data collection device for an electronic book, wherein the reading log collection unit collects a browsing target part at the time when the instruction is given as a reading log indicating a browsing interruption part for the electronic book. In the trial reading data creation device according to any one of claims 4 to 6,
When the instruction input unit is instructed to put the “terminal device itself” or “electronic book browsing function of the terminal device” in a standby state or stopped state while browsing the electronic book, the selection of the electronic book is canceled. Recognizing that the instructions were given,
A reading data creation device for an electronic book, wherein the reading log collection unit collects a browsing target part at the time when the instruction is given as a reading log indicating a browsing interruption part for the electronic book. In the trial reading data creation device according to any one of claims 4 to 7,
The reading log collecting unit collects the reading time as the reading log together with the instruction to specify the browsing target part.
The interest distribution generation unit recognizes the display time in which the same browsing target part is continuously displayed from the reading log, and if the display time exceeds a predetermined allowable range, the browsing target part is An electronic book trial reading data creation device characterized in that it is handled as a browsing interruption part. In the trial reading data creation device according to any one of claims 1 to 3,
An electronic book sample reading data creation device, wherein the interest level distribution creation unit creates an interest level distribution by counting the cumulative number of displays for each part of a creation target book as a positive interest level. In the trial reading data creation device according to any one of claims 1 to 3,
The reading log collecting unit collects the reading time as the reading log together with the instruction to specify the browsing target part.
The interest level distribution creation unit recognizes the display time of each browsing target part from the reading log, and counts the integrated value of the display time for each part of the created target book as a positive interest level to create an interest level distribution An electronic book trial reading data creation device characterized by: In the sample reading data creation device according to claim 10,
The reading log storage unit stores the reading log separately for each user,
The interest level distribution creating unit normalizes the display time for each part for each user, and then counts the integrated value of the normalized display time as a positive interest level to create an interest level distribution. An electronic book trial reading data creation device. In the trial reading data creation device according to claim 10 or 11,
An interest level distribution creation unit, when a display time for which the same browsing target part is continuously displayed exceeds a predetermined allowable range, the display time is not counted as an integrated value. Sample reading data creation device. In the trial reading data creation device according to any one of claims 1 to 3,
The instruction input unit has a function of accepting an instruction input for adding auxiliary information to a specific portion of the selected electronic book,
The browsing control unit stores the auxiliary information and has a function of adding and displaying the auxiliary information when displaying the specific location on a display,
A reading log collection unit collects a reading log including an instruction to add the auxiliary information, and transmits the reading log to the server device via the network,
An electronic book trial-reading data creation device, wherein the interest level distribution creation unit creates an interest level distribution by counting the addition frequency of the auxiliary information for each part of the creation target book as a positive interest level. In the trial reading data creation device according to claim 13,
A trial-reading data creation device for an electronic book, characterized by adding various marks, underlines, or memos as auxiliary information. In the trial reading data creation device according to any one of claims 1 to 14,
A trial reading data creation unit sets a division position at an arbitrary position in one high interest area, and uses data from the top of the high interest area to the division position as trial reading data. Sample reading data creation device. In the trial reading data creation device according to any one of claims 1 to 14,
The trial reading data creation unit sets a division position at an arbitrary position in one high interest area, and uses data from the division position to a position that is traced by a predetermined amount as trial reading data. Data creation device. In the trial reading data creation device according to any one of claims 1 to 14,
The trial reading data creation unit sets a division position at an arbitrary position in one high interest area, and data from the division position to a position that goes back to the beginning of the paragraph, section, chapter, or entire book including the division position A trial-reading data creation device for an electronic book, characterized in that the trial-reading data is used. In the trial reading data creation device according to any one of claims 1 to 14,
The trial reading data creation unit sets a division position at an arbitrary position in each high interest level area, divides each high interest level area into a first half part and a second half part with the set division position as a boundary, and distribution target data A trial-reading data creation device for electronic books, characterized in that data obtained by removing the latter half of each of the high interest areas from the electronic book data of one book stored in the storage unit is used as trial-reading data. A method for creating sample reading data for an electronic book provided via a network,
Based on the user's instruction input, the terminal device having a function of displaying the contents of the electronic book on the display and allowing the user to browse the contents collects the user's browsing history as a reading log, and this is collected via the network. A reading log sending stage to send,
A reading log reception stage in which a server device that creates trial reading data receives a reading log transmitted from a plurality of terminal devices via a network;
The server device analyzes the received reading log to obtain an interest level indicating the degree of interest of the user for each part of the specific electronic book, and creates an interest level distribution for the specific electronic book The degree distribution creation stage,
A high interest level region extraction step in which the server device sets a threshold value for the interest level of the interest level distribution and extracts a high interest level region where the interest level is equal to or higher than the threshold value;
Based on the electronic book data of the specific electronic book, the server device performs a trial reading by performing a process of extracting the first half of the data from the high interest area or a process of deleting the second half of the high interest area. The trial reading data creation stage to create data,
A test reading data creation method for an electronic book, comprising: In the sample reading data creation method according to claim 19,
By analyzing the reading log in the interest distribution creation stage, the user can recognize the browsing interruption part that is estimated to have interrupted the browsing, and for each part of the electronic book, the cumulative number of interruptions that became the browsing interruption part A trial reading data creation method for an electronic book, characterized in that the interest distribution is created by counting as a negative interest level. In the sample reading data creation method according to claim 19,
By analyzing the reading log at the stage of creating the interest level distribution, the cumulative number of displays for each part, the accumulated value of the display time, or the frequency of auxiliary information added by the user is counted as a positive interest level, and the interest level distribution A method for creating sample reading data of an electronic book, characterized by creating a document.

Images